ptpd on 2.6.30.4

[Patrick Ohly]

On Mon, 2009-08-17 at 15:28 +0100, John W. Linville wrote:
> On Wed, Aug 12, 2009 at 03:25:05PM -0400, Christopher Johnston wrote:
> > Good to see there is a maintainer for ptpd now and a future for improvements
> > and features.  At my company we have a large ptpd installment on our systems
> > for very precise timing in our production server (no ntp clients).  The new
> > Intel 82576 add-in for hardware timestamping looks promising but
> > unfortunately I can't seem to keep the time stable using the "linux_hw"
> > option, time goes backwards/forwards very aggresively.

Does it fail even in the simplest case of just one master and one slave?
Do you see anything suspicious in the kernel messages or the ptpd
output?

I haven't tested this for a while because my main job (SyncEvolution)
keeps me busy 200% of my time. Sorry for the vague answer, but I'm
really not into it much right now. I hate to do a code dump and then run
away from it, which is why I still keep an eye on what's going on. I
just can't do as much about it as I'd like to :-/

> >  So a little
> > clarification around whether I am using it correctly would be good.  Using
> > the default mode works fine as this is what I am using on 100s of servers
> > without a problem.
> > CPPFLAGS="-DPTPD_DBGV -DHAVE_LINUX_NET_TSTAMP_H
> > -I/root/linux-2.6.30.4/include"
> > 
> > Here are the flags I am using at compilation time.
> > 
> > Let me know if you need anything else.
> 
> Chris,
> 
> Sorry for the delayed response -- I wanted to give more people a chance
> to subscribe to the list, and I took a little trip this weekend... :-)
> 
> FWIW, I'm fairly new to the code here.  Besides, I don't think I have
> any of that fancy Intel hardware.

Duly noted. I don't have a budget, but perhaps someone else has.

> Patrick Ohly from Intel is on the list now.  He implemented the HW
> timestamp code.  Perhaps he can comment?

I'd better get the people involved who have taken over the maintenance
of the code inside the kernel and driver.

-- 
Best Regards, Patrick Ohly

The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.